Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis

Hanyeoreum Bae; Miryeong Kwon; Donghyun Gouk; Sanghyun Han; Sungjoon Koh; Changrim Lee; Dongchul Park; Myoungsoo Jung

doi:10.1109/ICCD53106.2021.00057

Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis

Hanyeoreum Bae, Miryeong Kwon, Donghyun Gouk, Sanghyun Han, Sungjoon Koh, Changrim Lee, Dongchul Park, Myoungsoo Jung

School of Integrated Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We investigate runtime environment characteristics and explore the challenges of conventional in-memory graph processing. This system-level analysis includes empirical results and observations, which are opposite to the existing expectations of graph application users. Specifically, since raw graph data are not the same as the in-memory graph data, processing a billion-scale graph exhausts all system resources and makes the target system unavailable due to out-of-memory at runtime.To address a lack of memory space problem for big-scale graph analysis, we configure real persistent memory devices (PMEMs) with different operation modes and system software frameworks. In this work, we introduce PMEM to a representative in-memory graph system, Ligra, and perform an in-depth analysis uncovering the performance behaviors of different PMEM-applied in-memory graph systems. Based on our observations, we modify Ligra to improve the graph processing performance with a solid level of data persistence. Our evaluation results reveal that Ligra, with our simple modification, exhibits 4.41× and 3.01× better performance than the original Ligra running on a virtual memory expansion and conventional persistent memory, respectively.

Original language	English
Title of host publication	Proceedings - 2021 IEEE 39th International Conference on Computer Design, ICCD 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	316-320
Number of pages	5
ISBN (Electronic)	9781665432191
DOIs	https://doi.org/10.1109/ICCD53106.2021.00057
Publication status	Published - 2021
Event	39th IEEE International Conference on Computer Design, ICCD 2021 - Virtual, Online, United States Duration: 2021 Oct 24 → 2021 Oct 27

Publication series

Name	Proceedings - IEEE International Conference on Computer Design: VLSI in Computers and Processors
Volume	2021-October
ISSN (Print)	1063-6404

Conference

Conference	39th IEEE International Conference on Computer Design, ICCD 2021
Country/Territory	United States
City	Virtual, Online
Period	21/10/24 → 21/10/27

Bibliographical note

Funding Information:
VII. ACKNOWLEDGEMENT This research is mainly supported by NRF 2021R1AC4001773 and IITP 2021-0-00524. The work is also supported in part by S3RC Hynix Center, SK-Hynix (G01200477), ETRI (21ZS1300), and KAIST start-up package (G01190015). D. Park is funded by NRF 2020R1F1A1048485. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies. Myoungsoo Jung is the corresponding author.

Funding Information:
This research is mainly supported by NRF 2021R1AC4001773 and IITP 2021-0-00524. The work is also supported in part by S3RC Hynix Center, SKHynix (G01200477), ETRI (21ZS1300), and KAIST start-up package (G01190015). D. Park is funded by NRF 2020R1F1A1048485. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.

Publisher Copyright:
© 2021 IEEE.

All Science Journal Classification (ASJC) codes

Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/ICCD53106.2021.00057

Cite this

Bae, H., Kwon, M., Gouk, D., Han, S., Koh, S., Lee, C., Park, D., & Jung, M. (2021). Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis. In Proceedings - 2021 IEEE 39th International Conference on Computer Design, ICCD 2021 (pp. 316-320). (Proceedings - IEEE International Conference on Computer Design: VLSI in Computers and Processors; Vol. 2021-October). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCD53106.2021.00057

Bae, Hanyeoreum ; Kwon, Miryeong ; Gouk, Donghyun et al. / Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis. Proceedings - 2021 IEEE 39th International Conference on Computer Design, ICCD 2021. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 316-320 (Proceedings - IEEE International Conference on Computer Design: VLSI in Computers and Processors).

@inproceedings{d7c4983132154d889c6002a1e69f7edc,

title = "Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis",

abstract = "We investigate runtime environment characteristics and explore the challenges of conventional in-memory graph processing. This system-level analysis includes empirical results and observations, which are opposite to the existing expectations of graph application users. Specifically, since raw graph data are not the same as the in-memory graph data, processing a billion-scale graph exhausts all system resources and makes the target system unavailable due to out-of-memory at runtime.To address a lack of memory space problem for big-scale graph analysis, we configure real persistent memory devices (PMEMs) with different operation modes and system software frameworks. In this work, we introduce PMEM to a representative in-memory graph system, Ligra, and perform an in-depth analysis uncovering the performance behaviors of different PMEM-applied in-memory graph systems. Based on our observations, we modify Ligra to improve the graph processing performance with a solid level of data persistence. Our evaluation results reveal that Ligra, with our simple modification, exhibits 4.41× and 3.01× better performance than the original Ligra running on a virtual memory expansion and conventional persistent memory, respectively.",

author = "Hanyeoreum Bae and Miryeong Kwon and Donghyun Gouk and Sanghyun Han and Sungjoon Koh and Changrim Lee and Dongchul Park and Myoungsoo Jung",

note = "Funding Information: VII. ACKNOWLEDGEMENT This research is mainly supported by NRF 2021R1AC4001773 and IITP 2021-0-00524. The work is also supported in part by S3RC Hynix Center, SK-Hynix (G01200477), ETRI (21ZS1300), and KAIST start-up package (G01190015). D. Park is funded by NRF 2020R1F1A1048485. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies. Myoungsoo Jung is the corresponding author. Funding Information: This research is mainly supported by NRF 2021R1AC4001773 and IITP 2021-0-00524. The work is also supported in part by S3RC Hynix Center, SKHynix (G01200477), ETRI (21ZS1300), and KAIST start-up package (G01190015). D. Park is funded by NRF 2020R1F1A1048485. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies. Publisher Copyright: {\textcopyright} 2021 IEEE.; 39th IEEE International Conference on Computer Design, ICCD 2021 ; Conference date: 24-10-2021 Through 27-10-2021",

year = "2021",

doi = "10.1109/ICCD53106.2021.00057",

language = "English",

series = "Proceedings - IEEE International Conference on Computer Design: VLSI in Computers and Processors",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "316--320",

booktitle = "Proceedings - 2021 IEEE 39th International Conference on Computer Design, ICCD 2021",

address = "United States",

}

Bae, H, Kwon, M, Gouk, D, Han, S, Koh, S, Lee, C, Park, D & Jung, M 2021, Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis. in Proceedings - 2021 IEEE 39th International Conference on Computer Design, ICCD 2021. Proceedings - IEEE International Conference on Computer Design: VLSI in Computers and Processors, vol. 2021-October, Institute of Electrical and Electronics Engineers Inc., pp. 316-320, 39th IEEE International Conference on Computer Design, ICCD 2021, Virtual, Online, United States, 21/10/24. https://doi.org/10.1109/ICCD53106.2021.00057

Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis. / Bae, Hanyeoreum; Kwon, Miryeong; Gouk, Donghyun et al.

Proceedings - 2021 IEEE 39th International Conference on Computer Design, ICCD 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 316-320 (Proceedings - IEEE International Conference on Computer Design: VLSI in Computers and Processors; Vol. 2021-October).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis

AU - Bae, Hanyeoreum

AU - Kwon, Miryeong

AU - Gouk, Donghyun

AU - Han, Sanghyun

AU - Koh, Sungjoon

AU - Lee, Changrim

AU - Park, Dongchul

AU - Jung, Myoungsoo

N1 - Funding Information: VII. ACKNOWLEDGEMENT This research is mainly supported by NRF 2021R1AC4001773 and IITP 2021-0-00524. The work is also supported in part by S3RC Hynix Center, SK-Hynix (G01200477), ETRI (21ZS1300), and KAIST start-up package (G01190015). D. Park is funded by NRF 2020R1F1A1048485. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies. Myoungsoo Jung is the corresponding author. Funding Information: This research is mainly supported by NRF 2021R1AC4001773 and IITP 2021-0-00524. The work is also supported in part by S3RC Hynix Center, SKHynix (G01200477), ETRI (21ZS1300), and KAIST start-up package (G01190015). D. Park is funded by NRF 2020R1F1A1048485. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies. Publisher Copyright: © 2021 IEEE.

PY - 2021

Y1 - 2021

N2 - We investigate runtime environment characteristics and explore the challenges of conventional in-memory graph processing. This system-level analysis includes empirical results and observations, which are opposite to the existing expectations of graph application users. Specifically, since raw graph data are not the same as the in-memory graph data, processing a billion-scale graph exhausts all system resources and makes the target system unavailable due to out-of-memory at runtime.To address a lack of memory space problem for big-scale graph analysis, we configure real persistent memory devices (PMEMs) with different operation modes and system software frameworks. In this work, we introduce PMEM to a representative in-memory graph system, Ligra, and perform an in-depth analysis uncovering the performance behaviors of different PMEM-applied in-memory graph systems. Based on our observations, we modify Ligra to improve the graph processing performance with a solid level of data persistence. Our evaluation results reveal that Ligra, with our simple modification, exhibits 4.41× and 3.01× better performance than the original Ligra running on a virtual memory expansion and conventional persistent memory, respectively.

AB - We investigate runtime environment characteristics and explore the challenges of conventional in-memory graph processing. This system-level analysis includes empirical results and observations, which are opposite to the existing expectations of graph application users. Specifically, since raw graph data are not the same as the in-memory graph data, processing a billion-scale graph exhausts all system resources and makes the target system unavailable due to out-of-memory at runtime.To address a lack of memory space problem for big-scale graph analysis, we configure real persistent memory devices (PMEMs) with different operation modes and system software frameworks. In this work, we introduce PMEM to a representative in-memory graph system, Ligra, and perform an in-depth analysis uncovering the performance behaviors of different PMEM-applied in-memory graph systems. Based on our observations, we modify Ligra to improve the graph processing performance with a solid level of data persistence. Our evaluation results reveal that Ligra, with our simple modification, exhibits 4.41× and 3.01× better performance than the original Ligra running on a virtual memory expansion and conventional persistent memory, respectively.

UR - http://www.scopus.com/inward/record.url?scp=85123941556&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85123941556&partnerID=8YFLogxK

U2 - 10.1109/ICCD53106.2021.00057

DO - 10.1109/ICCD53106.2021.00057

M3 - Conference contribution

AN - SCOPUS:85123941556

T3 - Proceedings - IEEE International Conference on Computer Design: VLSI in Computers and Processors

SP - 316

EP - 320

BT - Proceedings - 2021 IEEE 39th International Conference on Computer Design, ICCD 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 39th IEEE International Conference on Computer Design, ICCD 2021

Y2 - 24 October 2021 through 27 October 2021

ER -

Bae H, Kwon M, Gouk D, Han S, Koh S, Lee C et al. Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis. In Proceedings - 2021 IEEE 39th International Conference on Computer Design, ICCD 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 316-320. (Proceedings - IEEE International Conference on Computer Design: VLSI in Computers and Processors). doi: 10.1109/ICCD53106.2021.00057

Empirical Guide to Use of Persistent Memory for Large-Scale In-Memory Graph Analysis

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this